Reorder lto options in profiles.md (#15841)

Aside from `false`, the `lto` options seem to be ordered from "most
optimizing" to "least optimizing".

If this interpretation is correct, then I think `false` should go
between `thin` and `off`.

cc: @ehuss (who I think wrote that text)
This commit is contained in:
Weihang Lo 2025-08-17 15:36:33 +00:00 committed by GitHub
commit eae8d0dd30
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194

View File

@ -166,16 +166,16 @@ The `lto` setting controls `rustc`'s [`-C lto`], [`-C linker-plugin-lto`], and
LTO can produce better optimized code, using whole-program analysis, at the cost
of longer linking time.
The valid options are:
The valid options from most to least optimizing are:
* `false`: Performs "thin local LTO" which performs "thin" LTO on the local
crate only across its [codegen units](#codegen-units). No LTO is performed
if codegen units is 1 or [opt-level](#opt-level) is 0.
* `true` or `"fat"`: Performs "fat" LTO which attempts to perform
optimizations across all crates within the dependency graph.
* `"thin"`: Performs ["thin" LTO]. This is similar to "fat", but takes
substantially less time to run while still achieving performance gains
similar to "fat".
* `false`: Performs "thin local LTO" which performs "thin" LTO on the local
crate only across its [codegen units](#codegen-units). No LTO is performed
if codegen units is 1 or [opt-level](#opt-level) is 0.
* `"off"`: Disables LTO.
See the [linker-plugin-lto chapter] if you are interested in cross-language LTO.